Speech Reconstruction on English speech
3.92UTMOSWavTokenizer
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| WavTokenizerToken rate=75 Hz, BitRate (kbps)=0.92026.01 | 3.92 | 3.23 | 84 | |
| SACfs=12.5 Hz, fa=50 Hz, BitRate (kbps)=0.8752026.01 | 3.88 | 2.03 | 83 | |
| DualCodecsemantic_rate=12.5 Hz, Layers=6-layer, BitRate (kbps)=0.9252026.01 | 3.78 | 2.58 | 76 | |
| DSA-Tokenizerfs=25 Hz, fa=50 Hz, FSQ_levels_channels=4^8, Stacked_Layers=1, BitRate (kbps)=1.12026.01 | 3.67 | 5.93 | 60 | |
| DSA-Tokenizerfs=25 Hz, fa=25 Hz, FSQ_levels_channels=4^8, Stacked_Layers=1, BitRate (kbps)=0.72026.01 | 3.58 | 6.67 | 57 | |
| DSA-Tokenizerfs=25 Hz, fa=25 Hz, FSQ_levels_channels=4^7, Stacked_Layers=2, BitRate (kbps)=12026.01 | 3.56 | 8.77 | 60 | |
| DSA-Tokenizerfs=25 Hz, fa=25 Hz, FSQ_levels_channels=4^8, Stacked_Layers=2, BitRate (kbps)=1.12026.01 | 3.54 | 8.41 | 61 | |
| DSA-Tokenizerfs=25 Hz, fa=25 Hz, FSQ_levels_channels=4^8, Stacked_Layers=2, BitRate (kbps)=1.12026.01 | 3.46 | 2.61 | 77 | |
| DSA-Tokenizerfs=25 Hz, fa=25 Hz, FSQ_levels_channels=4^7, Stacked_Layers=2, BitRate (kbps)=12026.01 | 3.44 | 2.45 | 76 | |
| DSA-Tokenizerfs=25 Hz, fa=25 Hz, FSQ_levels_channels=4^8, Stacked_Layers=1, BitRate (kbps)=0.72026.01 | 3.41 | 2.09 | 72 | |
| DSA-Tokenizerfs=25 Hz, fa=50 Hz, FSQ_levels_channels=4^8, Stacked_Layers=1, BitRate (kbps)=1.12026.01 | 3.38 | 2.49 | 76 | |
| MimiLayers=8-layer, BitRate (kbps)=1.12026.01 | 3.3 | 3.46 | 74 | |
| DualCodecsemantic_rate=12.5 Hz, Layers=6-layer, BitRate (kbps)=0.9252026.01 | 2.39 | 17.36 | 11 | |
| MimiLayers=8-layer, BitRate (kbps)=1.12026.01 | 2.19 | 107.51 | 53 | |
| SpeechTokenizerLayers=2-layer, BitRate (kbps)=12026.01 | 2.12 | 7.57 | 36 | |
| EncodecLayers=2-layer, BitRate (kbps)=1.52026.01 | 1.56 | 5.4 | 61 | |
| SpeechTokenizerLayers=2-layer, BitRate (kbps)=12026.01 | 1.4 | 12.98 | 13 | |
| SACfs=12.5 Hz, fa=50 Hz, BitRate (kbps)=0.8752026.01 | 1.34 | 90.22 | 13 | |
| EncodecLayers=2-layer, BitRate (kbps)=1.52026.01 | 1.25 | 98.31 | 9 |